Summary: KLF5 is a basic transcription factor abundantly expressed in epithelial cells. It has a bidirectional role in cell proliferation control both promoting and inhibiting cell proliferation even in the same cells depending on growth factors. Whether and how such a bidirectional role occurs in vivo is unknown. KLF5 also has opposing functions in tumorigenesis, possessing both tumor suppressive and tumor promoter activities. How KLF5 executes opposing functions in tumorigenesis is another important question to address. In exploring how KLF5 functions in an unusual opposite manner, we found that in an in-vitro model of epithelial homeostasis, KLF5 is unacetylated and pro-proliferative without TGF-?, but becomes acetylated to inhibit cell proliferation when TGF-? is present. The mechanism for this functional switch is that the unacetylated pro-proliferative KLF5 transcriptional complex is reassembled upon TGF-?-induced KLF5 acetylation to alter gene regulation and cell proliferation control. In human tumors, Ac-KLF5 is downregulated and unAc-KLF5 is upregulated. We thus hypothesize that during epithelial homeostasis in adult prostates, KLF5 has a dual role that is determined by its acetylation status. Without differentiation signal, KLF5 is unacetylated and essential for progenitor cells to proliferate. When differentiation signal is activated, KLF5 becomes acetylated to alter gene transcription and initiate cell differentiation Insufficiency and deficiency of KLF5 acetylation disrupt cell differentiation and epithelial homeostasis, causing prostate cancer. We plan to test this hypothesis in three specific aims. (1) Expression of Klf5 in different types of cells and its bidirectional role in epithelial homeostasisin the prostate. Histological and molecular approaches will be applied to normal and genetically modified mice to examine Klf5 expression in different types of prostatic epithelial cells and address whether insufficiency (knockout) and deficiency (knockin) in Klf5 acetylation disrupt both proliferation and differentiation. (2) Opposite functions of Klf5 in prostate tumor initiation and progression. Klf5 will be knocked out in prostate epithelial cells. It will also be replaced wih acetylation-deficient Klf5K358R mutant (knockin). We will then test whether insufficiency (knockout) and deficiency (knockin) in Klf5 acetylation impair and abolish respectively Klf5's tumor suppressor function. Acetylation deficiency will be also tested in xenograft models of human prostate cancer. (3) How KLF5 executes opposing functions in epithelial homeostasis and tumorigenesis. Expression profiling combined with ChIP-Seq will be conducted to identify genes and gene signatures that are associated with KLF5's bidirectional function and acetylated and unacetylated KLF5. Such gene signatures will be compared with those of TGF-? and Pten from literature to evaluate their interactions with KLF5. We will also test whether and how excess growth signaling interferes with the acetylation of KLF5 when KLF5 is not deleted, which could lead to deacetylation of KLF5 and thus make KLF5 oncogenic. Genes mediating KLF5's oncogenic function have the potential to become biomarkers and therapeutic targets.